python - 替换 pandas 列中满足某些条件的值会导致SettingWithCopyWarning

Question

让dtrain属于 <class 'pandas.core.frame.DataFrame'> 类型

执行以下操作的正确方法是什么？

target = dtrain.iloc[:,1] > 0
dtrain.ix[target, 1] = 0

我收到警告:

/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/pandas/core/indexing.py:477: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead

注意: 早些时候我创建了 dtrain使用

dtrain = d.loc[(d.yyyy < 2005) & (d.yyyy >= 1995),:]

这可能是一个棘手的问题。我对 Python 相当陌生，对于什么触发深度复制以及什么只是指针，我的头脑中还没有一个好的模型......

Answer 1

我认为首先可以简化代码:

dtrain = d.loc[(d.yyyy < 2005) & (d.yyyy >= 1995),:]

至:

dtrain = d[(d.yyyy < 2005) & (d.yyyy >= 1995)]

它的名字是boolean indexing .

然后添加 copy -see docs .

dtrain = d[(d.yyyy < 2005) & (d.yyyy >= 1995)].copy()